Re: [Scikit-learn-general] Text Documents - Vectorizer

Le 30 mars 2012 14:50, Philipp Singer a écrit : > > I just have another question regarding this because some of my coworkers > brought this idea up and I cant argue about it the way I like. > > So let's assume you have 10 documents in the training test set and 10 > documents in the test set. > > M

Re: [Scikit-learn-general] GSoC 2012 pre-application

Le 30 mars 2012 07:19, Gael Varoquaux a écrit : > Hi Lee, > > Welcome! Thanks for preparing a proposal. My impression looking at it, is > that it seems a bit light for 2.5 months of work. It is pretty much > centered around implementing one algorithm, weighted k-means. One way to complement this

Re: [Scikit-learn-general] euro scipy

Le 30 mars 2012 13:04, Nelle Varoquaux a écrit : > > > I'll be there too, but I'm not sure if I'll be able to make it for a sprint. > > Also, pyconfr well be held mid september (15th - 16th) in Paris, and will > have rooms available on the 13th and 14th for sprints. Are people interested > ? I am

Re: [Scikit-learn-general] Text Documents - Vectorizer

Am 23.03.2012 13:58, schrieb Olivier Grisel: > Le 23 mars 2012 13:27, Philipp Singer a écrit : >> The IDF statistics is computed once on the whole training corpus as >> passed to the `fit` method and then reused on each call to the >> `transform` method. >> >> For a train / test split on typically

Re: [Scikit-learn-general] GSoC 2012 pre-application

Our affinity propagation really doesn't scale. I quickly tried to make it work on sparse matrices, but it turned out to be some work. That said, it's a crappy algorithm :$. Gael - Original message - > Hi Lee. > I'd have to have a look that the papers again to judge this better. > Maybe

Re: [Scikit-learn-general] GSoC 2012 pre-application

Hi Lee. I'd have to have a look that the papers again to judge this better. Maybe I'll have time on the weekend. What I would also like to see in the clustering module would be a more scalable mean shift and maybe also quickshift. Cheers, Andy On 03/30/2012 06:30 AM, Lee Zamparo wrote: > Hello

Re: [Scikit-learn-general] euro scipy

On 03/30/2012 01:04 PM, Nelle Varoquaux wrote: On 30 March 2012 12:04, Olivier Grisel > wrote: Le 30 mars 2012 10:37, Gilles Louppe mailto:g.lou...@gmail.com>> a écrit : > Since it's in Brussels, I think I should be there as well :) > > I can a

Re: [Scikit-learn-general] euro scipy

On 30 March 2012 12:04, Olivier Grisel wrote: > Le 30 mars 2012 10:37, Gilles Louppe a écrit : > > Since it's in Brussels, I think I should be there as well :) > > > > I can also help with something around scikit-learn if needed. > > I will be there as well. > > Would be great to do a sprint ind

Re: [Scikit-learn-general] Coordinated descent in linear models beyond squared loss GSOC

On Fri, Mar 30, 2012 at 6:20 PM, Gael Varoquaux < gael.varoqu...@normalesup.org> wrote: > One remark on the algorithmic side, rather than on the project management > side: it seemed to me when I read the Yuan paper than the strategy to > choose the order in which the variables where updated could

Re: [Scikit-learn-general] GSoC 2012 pre-application

Regarding clustering algorithms, I would suggest to have a look at convex formulations, that can be advantageous for the sake of convergence/stability, wrt standard algorithms that never have any guarantee. Among others: - http://www.icml-2011.org/papers/419_icmlpaper.pdf - http://www.google.fr

Re: [Scikit-learn-general] euro scipy

Le 30 mars 2012 10:37, Gilles Louppe a écrit : > Since it's in Brussels, I think I should be there as well :) > > I can also help with something around scikit-learn if needed. I will be there as well. Would be great to do a sprint indeed. -- Olivier http://twitter.com/ogrisel - http://github.c

Re: [Scikit-learn-general] Coordinated descent in linear models beyond squared loss GSOC

Hi Immanuel, Your proposal looks interesting. It is really welcomed. I like the list of strategies for optimization that you have given. One remark on the algorithmic side, rather than on the project management side: it seemed to me when I read the Yuan paper than the strategy to choose the order

Re: [Scikit-learn-general] euro scipy

Since it's in Brussels, I think I should be there as well :) I can also help with something around scikit-learn if needed. Gilles On 30 March 2012 10:31, Vincent Michel wrote: > I think that I will be there too. > > > 2012/3/30 Alexandre Gramfort >> >> I should be there. >> >> I could help wit

Re: [Scikit-learn-general] sklearn.hmm ownership?

On Thu, Mar 29, 2012 at 17:43, Daniel Duckworth wrote: > Has someone taken ownership of sklearn.hmm?  If the task is open or help > would be appreciated, I'd like to jump in. That's great. I know I have no idea why it has some of the bugs it has, which I tripped all over while trying to implement

Re: [Scikit-learn-general] euro scipy

I think that I will be there too. 2012/3/30 Alexandre Gramfort > I should be there. > > I could help with something around the scikit. > > Alex > > On Fri, Mar 30, 2012 at 1:50 AM, Gael Varoquaux > wrote: > > On Fri, Mar 30, 2012 at 01:30:47AM +0200, Andreas Mueller wrote: > >> I just wanted to

Re: [Scikit-learn-general] euro scipy

I should be there. I could help with something around the scikit. Alex On Fri, Mar 30, 2012 at 1:50 AM, Gael Varoquaux wrote: > On Fri, Mar 30, 2012 at 01:30:47AM +0200, Andreas Mueller wrote: >> I just wanted to ask who of you is going to euroscipy. > > I am, and I few people from our group. >